OK, the title is a little ridiculous. We actually debated changing the timeframe just so that it didn’t sound way too sensational, but we decided not to because it wouldn’t be honest. So hold your skepticism, or don’t, but at least read the article until the end.

TLDR: A new campaign structure was put into place on Facebook and Instagram ads that immediately had a dramatic effect in reversing this trailing trend. The philosophy behind the structure lies in creating two funnels, one for users who’ve never purchased the product, and one for existing customers. This allows us to easily prioritize going after the most incremental purchases with separate goals and fluctuate dollars between these groups depending on what we’re seeing happen organically. We see more actual profit being generated, while the amount of revenue the platform is able to take credit for (or attribute) may even decrease.

The Challenge:  

Our client, a children’s apparel store, came to us with declining net new customer growth year over year for the past 8 months. Going into the busiest quarter for ecommerce, they knew they had to make a change to remain a top competitor in their market.  

Then our client learned about Elumynt’s approach to paid media: Focus on the overall business’s EBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization aka profit) rather than what the platforms are reporting for ROAS.

Sidenote: If you don’t know how Meta’s reporting is inaccurate, you’re way behind. Aggregated Data Dumping has become more and more prevalent. In-platform conversion data—especially at the ad level—is now unreliable for determining what levers to pull.

The Strategy:

Elumynt is an ecommerce performance marketing agency focused on rapid and sustainable growth, with an emphasis on new customer acquisition and profits. The way Meta advertising works (in theory) is it serves your ads to the people it thinks are most likely to convert. And who’s most likely to convert? Your existing customers! Plus, users who are familiar with your brand, have engaged with your brand, have something in their cart on your website, etc. But if you’re looking to GROW, then you need to force Meta to look BEYOND these users.

Time and time again, when we audit a business’s Facebook ad account, there are a lot of campaigns labeled ”Prospecting”. When you think of prospecting, how do you describe it? Most of the people we polled said something along the lines of “trying to get someone who’s never heard of you to be aware of your brand or buy from your brand.” Even when people understand the logic, we rarely come across an ad account where they’re excluding even their first-party data from the audience pool. So then it goes like this…  

So, how did we help them turn around their YoY customer growth in one month?  

We implemented our bread-and-butter – a strict net new customer vs. existing customer campaign structure (if you’re interested in the nitty-gritty of exactly how we set this up, you can scroll down to the bottom). We focused a majority of spend on new customer acquisition. Don’t worry, we didn’t neglect the current customer base. We had different CPA goals for these users and spent significantly less on existing customers since our client has had fantastic success turning their new fans into repeat customers through low-cost SMS and email marketing.  

The other important part of our strategy was using a click-based attribution window (for optimization, not necessarily for reporting) to give Facebook positive reinforcement for only the most incremental conversions it drove.  

We then optimized channels against EBITDA (Earnings Before Interest Taxes, Deductions, and Appreciation – essentially “profit”) and measured CAC constantly to grow true profit and growth, rather than optimizing to an arbitrary ROAS number that simply ‘feels good’.

Outside our major structural changes in platform, we shifted the way the client was evaluating success. Instead of focusing on in-platform efficiency and attribution we went for the big goals that truly move the business forward. We gathered business data such as Cost of Goods Sold (COGS), shipping, overhead, etc, to understand what spend to revenue ratio (MER) they needed to be at to drive profitable growth. And as we always do, we took it a step further and looked at a multitude of spend and MER models to not only be profitable but to scale EBITDA YoY for this business.

The (Killer) Results

The overall impact was massive. After months of new customer growth declining YoY, we were able to significantly turn that around during the most competitive month in ecommerce. New customer orders were down 9% YoY from August to October of 2021. When Elumynt restructured their Facebook ad account in late October, orders skyrocketed, increasing 29% YoY from November through January. It’s also important to note they had not seen this level of customer growth in November in years past.

Don’t let inaccurate ROAS or lack of audience segmentation strategy limit your usage of paid social in your business’s growth

Because of this significant increase in new customer growth, we were able to reassure our client there is plenty of room for them to grow via Facebook and Instagram ads, which they previously had lost confidence in due to the diminishing returns they had seen when trying to scale their previous structure.

Our work isn’t over, and we continue to find new ways to unlock opportunities and allow the platform to work better for our needs, but the core of this philosophy is our biggest key to success, and now you have it too 😊  

Also, if you’re curious… this brand is still growing strong, and according to Varos (a Y-combinator-backed, relative benchmarking tool for ecommerce) suggests that Elumynt is so far beyond the top quartile, it’s not even funny.

The Nitty Gritty

To make this distinction between campaigns, prospecting is easy. For going after new customers, we exclude 180 day purchasers based on the pixel (as far back as Meta allows), along with every other event we’re prioritizing in Aggregated Events Measurement, since opt-out users only get the highest prioritized event tracked (if we had prioritized events: View Content, Add to Cart, Purchase, and we were only excluding users who’ve viewed content, then we wouldn’t be excluding opt out users who added to cart or purchased since that would be the only tracked event for them). We also exclude all first party data via an auto-syncing Klaviyo integration and anyone who has clicked a Call to Action (CTA) on a Facebook ad or engaged with the client’s Instagram profile in the last 365 days.  

Then, to serve ads to existing customers to get them to buy again, we target the client’s full email list (via Klaviyo and the pixel) while excluding anyone who’s purchased in the last month, as they’re likely not ready to buy again yet. We can have a lot of fun with this by segmenting out one-time buyers, lapsed buyers, non-subscribers, etc.  

Retargeting gets tricky, and the include/exclude logic has gotten a bit messier since iOS14, so I won’t bore you with the further nitty gritty, but rest assured, we segment those out, too.

Admittedly, there’s still a LOT that hasn’t been covered here – this is the ~10% that “the boss” was willing to let me share publicly right now. The other 90% of what we’re doing to dramatically scale brands you’ll have to pay to play for if you want to see what turns this strategy into pure fire. Don’t shoot the messenger 🤷🏻‍♀️

‍